Uncertainties in estimating the effect of climate change on 100-year return value for significant wave height

TL;DR

This study assesses the uncertainties in estimating 100-year wave height return values due to climate change, using Bayesian extreme value analysis on climate model outputs across different regions, highlighting large model-related variability.

Contribution

It introduces a Bayesian framework for non-stationary extreme value analysis of wave heights, quantifies uncertainties across multiple climate models, and provides software tools for future assessments.

Findings

Significant variability in return value estimates across models and locations.

Model choice and regional storm characteristics greatly influence uncertainty.

Long-term climate model data helps quantify inherent variability in extreme wave estimates.

Abstract

Estimating climate effects on future ocean storm severity is plagued by large uncertainties, yet for safe design and operation of offshore structures, best possible estimates of climate effects are required given available data. We explore the variability in estimates of 100-year return value of significant wave height (Hs) over time, for output of WAVEWATCHIII models from 7 representative CMIP5 GCMs, and the FIO-ESM v2.0 CMIP6 GCM, for neighbourhoods of locations east of Madagascar and south of Australia. Non-stationary extreme value analysis of peaks-over-threshold and block maxima using Bayesian inference provide posterior estimates of return values as a function of time; MATLAB software is provided. There is large variation between return value estimates from different GCMs, and with longitude and latitude within each neighbourhood. These sources of uncertainty tend to be larger than that due to typical modelling choices (such as choice of threshold for POT, or block length for BM). However, careful threshold and block length are critical east of Madagascar, because of the presence of a mixed population of storms there. The long 700-year pre-industrial control (piControl) output of the CMIP6 GCM allows quantification of the apparent inherent variability in return value as a function of time.